# US 10,500,721 B2 · Machine learning device, laminated core manufacturing apparatus, laminated core manufacturing system, and machine learning method for learning operation for stacking core sheets > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,500,721 B2 - Document ID: US10500721B2 - Status: Granted - Kind: B2 - Application number: 15/368,753 - Filed: 2016-12-05 - Published: 2017-06-08 - Granted: 2019-12-10 - Earliest claimed: 2015-12-07 - Assignees: Fanuc Corporation - Applicants: Fanuc Corporation - Inventors: Suguru Takamizawa - Web page: https://patentgrep.com/patent/US10500721B2 ## CPC - B25J 9/163: {learning, adaptive, model based, rule based expert control} - B25J 9/1687: {Assembly, peg and hole, palletising, straight line, weaving pattern movement} - B25J 9/1697: {Vision controlled systems} - G06N 3/0499: Feedforward networks - G06N 3/08: Learning methods - G06N 3/0895: Weakly supervised learning, e.g. semi-supervised or self-supervised learning - G06N 3/09: Supervised learning - G06N 3/092: Reinforcement learning - G06N 3/098: Distributed learning, e.g. federated learning - H01F 41/0233: {Manufacturing of magnetic circuits made from sheets (magnetic cores made from sheets H01F27/245; soft magnetic alloys in the form of sheets H01F1/16)} - G05B 2219/39271: Ann artificial neural network, ffw-nn, feedforward neural network - G05B 2219/45055: Assembly - Y10S 901/03: Teaching system - B25J: MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES ({manipulators specially adapted for use in surgery A61B34/70; manipulators used in cleaning hollow articles B08B9/04}; manipulators associated with rolling mills B21B39/20; manipulators associated with forging machines B21J13/10; {manipulators associated with picking-up and placing mechanisms B23P19/007}; means for holding wheels or parts thereof B60B30/00; {vehicles with ground-engaging propulsion means, e.g. walking members B62D57/02, B62D57/032; devices for picking-up and depositing articles or materials between conveyors B65G47/90, B65G47/91; manipulators with gripping or holding means for transferring packages B65H67/065}; cranes B66C; {manipulators used in the protection or supervision of pipe-line installations F17D5/00; walking equipment adapted for nuclear steam-generators F22B37/006}; manipulators specially adapted for, or associated with, nuclear reactors G21C; {apparatus used for handling wafers during manufacture or treatment of semiconductor H10P72/50}) - G06N: COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS - H01F: MAGNETS; INDUCTANCES; TRANSFORMERS; SELECTION OF MATERIALS FOR THEIR MAGNETIC PROPERTIES - G05B: CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS (systems for controlling or regulating non-electric variables G05D; systems for regulating electric or magnetic variables G05F; control devices or systems insofar as characterised by mechanical features only G05G) - Y10S: TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS ## Abstract A machine learning device which learns an operation of a laminated core manufacturing apparatus for stacking a plurality of core sheets to manufacture a laminated core, wherein the machine learning device includes a state observation unit which observes states of the core sheets and the laminated core manufacturing apparatus; and a learning unit which updates a manipulated variable for stacking the core sheets, on the basis of a state variable observed by the state observation unit. ## Claims ### Claim 1 A laminated core manufacturing apparatus, comprising: a robot having a hand portion configured to stack core sheets having a predetermined shape one by one or in groups of at least two core sheets to manufacture a laminated core; and a robot vision system attached to the hand portion, wherein the laminated core manufacturing apparatus is configured to manufacture the laminated core by stacking the core sheets based on (1) shape data of the core sheets detected by the robot vision system, (2) operation data of the laminated core manufacturing apparatus, and (3) time data for stacking the core sheets. ### Claim 2 (depends on 1) The laminated core manufacturing apparatus according to claim 1, further comprising: a machine learning device configured to learn an operation of the laminated core manufacturing apparatus, the machine learning device comprising: a state observation unit configured to observe states of the core sheets and the laminated core manufacturing apparatus; and a learning unit configured to update a manipulated variable for stacking the core sheets, on the basis of a state variable observed by the state observation unit, the state variable representing (1) the shape data of the core sheets, (2) the operation data of the laminated core manufacturing apparatus, and (3) the time data for stacking the core sheets. ### Claim 3 (depends on 2) A laminated core manufacturing system comprising a plurality of laminated core manufacturing apparatuses according to claim 2, wherein the machine learning device is mounted in each of the laminated core manufacturing apparatuses, and the plurality of machine learning devices mounted in the plurality of laminated core manufacturing apparatuses, respectively, are configured to share or exchange data with one another via a communication medium. ### Claim 4 (depends on 2) A laminated core manufacturing system comprising a plurality of laminated core manufacturing apparatuses according to claim 2, wherein the machine learning device of at least one of the plurality of laminated core manufacturing apparatuses is located on a cloud server. ### Claim 5 A machine learning method of learning an operation of a laminated core manufacturing apparatus for stacking a plurality of core sheets to manufacture a laminated core, the method comprising: observing states of the core sheets and the laminated core manufacturing apparatus; and updating a manipulated variable for stacking the core sheets, on the basis of a state variable obtained in the observing, the state variable representing (1) shape data of the core sheets, (2) operation data of the laminated core manufacturing apparatus, and (3) time data for stacking the core sheets, wherein the laminated core manufacturing apparatus comprises a robot comprising a robot vision system attached to one of a hand portion and an area which provides a broad view of a workspace, and in said observing, the shape data of the core sheets are received from the robot vision system. ### Claim 6 A machine learning device for learning an operation of a laminated core manufacturing apparatus for stacking a plurality of core sheets to manufacture a laminated core, wherein the machine learning device comprises: a state observation unit configured to observe states of the core sheets and the laminated core manufacturing apparatus; and a learning unit configured to update a manipulated variable for stacking the core sheets, on the basis of a state variable observed by the state observation unit, the state variable representing (1) shape data of the core sheets, (2) operation data of the laminated core manufacturing apparatus, and (3) time data for stacking the core sheets, wherein the laminated core manufacturing apparatus comprises a robot comprising a robot vision system attached to one of a hand portion and an area which provides a broad view of a workspace, and the state observation unit is configured to receive the shape data of the core sheets from the robot vision system. ### Claim 7 (depends on 6) The machine learning device according to claim 6, wherein the laminated core manufacturing apparatus further comprises a robot control unit configured to issue a command of one of a displacement and an angle of the hand portion to control an operation of the robot, and the state observation unit is configured to receive from the robot control unit, the operation data of the laminated core manufacturing apparatus, and the time data for stacking the core sheets. ### Claim 8 (depends on 7) The machine learning device according to claim 7, wherein the state observation unit is configured to obtain a velocity gain and a position gain of the hand portion output from the robot control unit and the time data for stacking the core sheets. ### Claim 9 (depends on 6) The machine learning device according to claim 6, further comprising: a decision unit configured to decide an operation of the laminated core manufacturing apparatus by referring to the manipulated variable learned by the learning unit. ### Claim 10 (depends on 6) The machine learning device according to claim 6, wherein the learning unit comprises: a reward computation unit configured to compute a reward on the basis of output from the state observation unit; and a value function update unit configured to update a value function for determining a value of the operation for stacking the core sheets to manufacture a laminated core, in accordance with the reward on the basis of outputs from the state observation unit and the reward computation unit. ### Claim 11 (depends on 6) The machine learning device according to claim 6, wherein the learning unit comprises: an error computation unit configured to compute an error on the basis of output from the state observation unit and input teacher data; and an error model update unit configured to update an error model for determining an error of the operation for stacking the core sheets to manufacture a laminated core, on the basis of outputs from the state observation unit and the error computation unit. ### Claim 12 (depends on 6) The machine learning device according to claim 6, wherein the machine learning device further comprises a neural network.